Seediscussions,stats,andauthorprofilesforthispublicationat:http://www.researchgate.net/publication/275275527

Contemporaryevolutionofaninvasiveplantisassociatedwithclimatebutnotwithherbivory

ARTICLEinFUNCTIONALECOLOGY·APRIL2015

ImpactFactor:4.86·DOI:10.1111/1365-2435.12463

DOWNLOADS

5

VIEWS

24

6AUTHORS,INCLUDING:

PaoloZuccarini

PlantandFoodResearch

21PUBLICATIONS114CITATIONS

SEEPROFILE

AnnaEscolà

AutonomousUniversityofBarcelona

3PUBLICATIONS0CITATIONS

SEEPROFILE

LauraArmengot

UniversityofBarcelona

16PUBLICATIONS93CITATIONS

SEEPROFILE

EvaCastells

AutonomousUniversityofBarcelona

21PUBLICATIONS186CITATIONS

SEEPROFILE

Availablefrom:PaoloZuccarini

Retrievedon:27June2015

Acc

epte

d A

rtic

le

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1365-2435.12463

This article is protected by copyright. All rights reserved.

Received Date : 16-Jun-2014

Revised Date : 25-Mar-2015

Accepted Date : 10-Apr-2015

Article type : Standard Paper

Editor : Edith Allen

Section : Evolutionary Ecology

Contemporary evolution of an invasive plant is associated with climate but not

with herbivory

Ferran Colomer-Ventura a,b, Jordi Martínez-Vilalta a,b, Paolo Zuccarini c, Anna Escolà

d, Laura Armengot e, Eva Castells d,*

a CREAF, Cerdanyola del Vallès, 08193 Barcelona, Spain

b Departament de Biologia Animal, Vegetal i Ecologia, Universitat Autònoma Barcelona, Cerdanyola del

Vallès, 08193 Barcelona, Spain

c IRTA, Caldes de Montbui, Barcelona, Spain

d Departament de Farmacologia, Terapèutica i Toxicologia, Universitat Autònoma de Barcelona, Cerdanyola

del Vallès, Catalonia, Spain

e Departament de Biologia Vegetal, Universitat de Barcelona, Barcelona, Spain

*Corresponding author: eva.castells@uab.cat, Fax. +34 935812959

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Running headline: Contemporary evolution of an invasive plant

Summary

1. Divergence in plant traits and trait plasticity after invasion has been proposed as mechanisms

favouring invasion success. Current hypotheses predict a rapid evolution in response to changes in the

abiotic conditions in the area of introduction or to differences in the herbivore consumption pressure

caused by a decrease in the enemies associated with the area of origin (e.g., evolution of increased

competitive ability –EICA– hypothesis). The importance of these factors in determining plant

geographical divergence has not been yet simultaneously evaluated.

2. Senecio pterophorus (Asteraceae) is a perennial shrub native to Eastern South Africa and a recent

invader in Western South Africa (since ~100 years ago), Australia (>70-100 years) and Europe (>30

years). These areas differ in their summer drought stress (measured as the ratio between summer

precipitation and potential evapotranspiration, P/PET) and their interactions with herbivores.

3. We performed a common garden experiment with S. pterophorus sampled throughout its entire known

distributional area to determine: 1) whether native and non-native populations diverge in their traits,

as well as the plasticity of these traits in response to water availability, and 2) whether climate and

herbivory play a role in the genetic differentiation across regions.

4. Plants from the non-native regions were smaller and had a lower reproductive output than plants from

the indigenous area. No geographical differences in phenotypic plasticity were found in response to

water availability. Herbivory was not related to the plant geographical divergence. In contrast, our

results are consistent with the role of climate as a driver for post-invasive evolution, as suggested by

adaptation of plants to a drought cline in the native range, the analogous change in plant traits in

independently invaded regions, and the convergence of vegetative traits between non-native plants

and native plants under similar drought conditions.

5. Native and non-native populations of S. pterophorus differed in plant traits, but not in trait plasticity,

in response to their local climatic conditions. Our results are contrary to the role of herbivory as a

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

selective factor after invasion and highlight the importance of climate driving rapid evolution of

exotic plants.

Key-words: adaptation, biological invasions, drought, ecological clines, evolution of increased

competitive ability (EICA) hypothesis, herbivory, invasion ecology, phenotypic plasticity, plant traits, Senecio

pterophorus

Introduction

The number of invasive plants has risen dramatically during the last decades, impacting the structure, function

and dynamics of the receiver ecosystems (Mack et al. 2000). Successful plant invaders that become

established and spread into a new habitat, however, represent only few of the overall introduced species

(Williamson 1996; Kolar & Lodge 2001). Understanding why some species become invasive and others do

not is essential to predicting the outcomes of future introductions (Sol et al. 2012) but it remains an open issue

in the study of biological invasions.

During the last decade, the rapid evolution of exotic plants has been proposed as an important

determinant for invasion success (Maron et al. 2004; Prentis et al. 2008; Buswell, Moles & Hartley 2011).

Evolutionary change to novel environmental conditions is revealed by the divergence in genetically

determined traits between the native and the invasive populations. It is commonly expected that plant

genotypes with morphological and physiological traits related to higher fitness, such as an elevated growth,

biomass, reproductive capacity and competitive ability, will increase their frequency in the newly established

populations as a result of natural selection (Crawley 1987; Richards et al. 2006; Lachmuth, Durka & Schurr

2011). The characterization of the traits related to invasion, however, has proven difficult, in part because

successful strategies may vary among ecosystem types and climatic conditions (Sakai et al. 2001; Pyšek &

Richardson 2007). For example, small plants with narrow and thicker leaves perform better, and thus may be

favourably selected, in warm, dry and nutrient-poor environments compared with large plants with a high

foliar area (Westoby et al. 2002; Moles et al. 2009; Buswell, Moles & Hartley 2011). Plants may locally adapt

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

to new climatic conditions after invasion, expressing those traits that confer them higher fitness in the invaded

area (Colautti & Barrett 2013).

Contemporary evolution of exotic populations could also be driven by changes in the plant biotic

environment. The tendency of exotic plants to perform better in the areas of introduction than in the native

areas, expressed by an increased biomass, reproductive effort or competitive ability, has been attributed to the

release from herbivorous natural enemies (Crawley, 1987; Blossey & Notzold 1995; Keane & Crawley 2002).

The most cited hypothesis predicting post-invasive evolution, the Evolution of Increased Competitive Ability

(EICA) hypothesis (Blossey & Notzold 1995) states that under a lower consumption pressure, genotypes

allocating more resources to growth and reproduction and less to chemical defences would be favoured over

the less-competitive and more heavily defended plants.

In addition to the adaptive changes of plant trait values driven by novel abiotic and biotic environmental

conditions, invasive potential may also be determined by changes in the plastic response of those traits

(Richards et al. 2006). A high phenotypic plasticity (i.e., the ability of an organism to express distinct

phenotypes depending on the environmental conditions) expands the ecological niche and facilitates the

colonization of novel habitats (Richards et al. 2006, Berg & Ellers 2010). Accordingly, invasive plants are

expected to evolve an elevated plasticity in comparison with plants from the habitat of origin (Bossdorf et al.

2005). The hypothesis of the evolution of increased phenotypic plasticity after invasion, however, has been

rarely tested (Bossdorf et al. 2005; Vanderhoeven et al. 2010; Godoy, Valladares & Castro-Díez 2011).

Here we study whether plant traits and trait plasticity rapidly evolve in response to new abiotic and biotic

environmental conditions, using Senecio pterophorus DC (Asteraceae) as a model species. S. pterophorus is a

perennial shrub native to Eastern South Africa that has expanded to Western South Africa (ca. 100 years ago)

and has been introduced in Australia (> 70-100 years ago) and Europe (> 30 years ago) (Castells et al. 2013).

These four regions differ in their climatic conditions and interactions with herbivores (Hijmans et al. 2005;

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Castells et al. 2013) (Table 1; see Fig. S1 in Supporting Information). The distribution of S. pterophorus in its

native range occurs along an ecological cline of drought, but on average its native range is characterized by

wetter and hotter summers compared with all non-native regions. A biogeographical study showed that non-

native plants were released from herbivory after introduction, and this release was more intense in Europe, the

region with a shorter time span since introduction (Castells et al. 2013). These differences make S.

pterophorus a suitable model species to test the simultaneous role of key abiotic and biotic factors as

determinants of plant geographical divergence.

We conducted a common garden experiment using 47 populations of S. pterophorus spanning its

entire known distributional area across the native (Eastern South Africa), the expanded (Western South

Africa) and two introduced ranges (Australia and Europe). We determined the genetic differences in

individual-level traits, leaf-level traits and reproductive traits across regions and their response to water

availability, and we asked two main questions: 1) Do plants from the expanded and introduced populations

diverge in their traits and the phenotypic plasticity of those traits in comparison with the native populations?

2) Are climate and herbivory driving this genetic differentiation across regions?

The reported differences in the abiotic and biotic environment across regions allow us to make predictions

about the factors determining plant adaptation. If climate is driving the biogeographical divergence in plant

traits, we would expect a lower growth and leaf area in the introduced areas, where the plants are subject to

drier conditions, compared with the native populations. Moreover, this pattern should be similar in all of the

introduced areas because they share a similar climate. In contrast, if plant traits are explained by differences in

herbivory among regions, the non-native plants should grow more and have a higher reproductive output

compared with the native plants, especially in Europe, where herbivore release has been more intense

(Castells et al. 2013). Finally, regardless of the factors driving post-invasive changes, plants from the non-

native populations are expected to show a higher phenotypic plasticity in response to an environmental stress

compared with the native populations.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Materials and methods

MODEL SPECIES

Senecio pterophorus (Asteraceae) is a perennial shrub of 0.4 to 2 m in height that colonizes open and

disturbed environments, such as grasslands, forest margins and roads (Parsons & Cuthbertson 1992; Castells

et al. 2013). S. pterophorus is native to the Natal province in Eastern Cape, South Africa and was introduced

into the Western Cape Province circa 1918 (Hilliard 1977). The first citation in Australia is from 1908, but the

species became invasive approximately 1930 along the southern coast (Parsons & Cuthbertson 1992). In

continental Europe, S. pterophorus was first found near Barcelona, NE Spain, in 1982 and later in Liguria,

NW Italy, in 1990 (see references in Castells et al. 2013). Since 1994, S. pterophorus has been considered a

noxious weed subject to eradication by the Department of Environment and Primary Industries, Victoria

(Australia), and it has been recently catalogued as an invasive species in Catalonia (NE Spain) (Andreu et al.

2012). A detailed species description, distribution and invasion history of S. pterophorus is provided in

Castells et al. (2013).

FIELD SAMPLING

Senecio pterophorus was collected in 2009 and 2010 from 47 populations across the native range (Eastern

Cape in South Africa), the expanded range (Western Cape in South Africa) and two invasive ranges (Australia

and Europe) (Castells et al. 2013) (Table 1). Populations were at least 30 km apart in South Africa and

Australia and 5 km apart in Europe, covering the entire species’ known range, including the distribution limits

(Castells et al. 2013). In each population we collected seeds from 6 to 15 individuals (referred to here as

mother plants).

We calculated the ratio between summer precipitation and potential evapotranspiration (P/PET)

(Thornthwaite 1948) for each population as a measure of drought stress. Mean temperature and precipitation

during summer (June to August in the Northern Hemisphere and December to February in the Southern

Hemisphere) were obtained from the WorldClim database (Hijmans et al. 2005). Summer P/PET was

preferred over latitude or other climatic variables because it better relates to plant drought stress (e.g.,

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Martínez-Vilalta et al. 2008). Populations in the native range had, on average, a higher summer P/PET (lower

drought stress) compared with the non-native populations (Table 1). Additionally, native populations showed

a latitudinal cline of summer P/PET, with an increasing summer P/PET (decreasing summer drought) towards

the north (see Table S1).

The intensity of herbivore consumption on reproductive parts (heads and seeds) was characterized at

the original sampling locations on the same individuals used in the common garden experiment. Native and

expanded populations in South Africa had higher predation levels compared with the cross-continental

introductions, with Europe showing nearly a complete release from herbivores (Castells et al. 2013) (Table 1).

COMMON GARDEN EXPERIMENT

Experimental design

In November 2010, seeds from six individuals from 47 populations (a total of 282 mother plants) were

germinated in a mixture of Sphagnum, perlite and vermiculite (2:1:1) at the greenhouse facilities of the

Faculty of Biology, University of Barcelona (Spain). Soil was watered regularly with a Hoagland nutrient

solution. In February 2011, when the seedlings had four to five true leaves (100 days old approximately), two

seedlings per mother plant were transplanted to the common garden. We ignored whether the seedlings from a

mother plant were full-sibs or half-sibs, but for the sake of simplicity we refer to them as half-sibs throughout

this study. The common garden was conducted at the experimental fields of the Autonomous University of

Barcelona (41°29'53.3’’N, 02°06'9.6’’E) located in an old cultivated area surrounded by a Pinus halepensis

forest. The soil is typic calcixerept (Soil Survey Staff 2010) and the mean annual temperature and

precipitation are 14.9 ºC and 562.8 mm, respectively (Ninyerola et al. 2003). The weather in 2011, when the

experiment was performed, was hotter and wetter than the average (15.6 ºC and 853.1 mm) (Meteorological

Service of Catalonia 2015).

The field was divided into six plots of 58 m2 separated by 1.5 m. Three plots were left without

irrigation but receiving rainfall (Not Watered, NW) and three plots were assigned to a drip irrigation treatment

(Watered, W) set at 4.5 L/day/plant. Treatments were randomly assigned to plots. Drought was selected as a

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

stress treatment to characterize phenotypic plasticity, as growth and survival of S. pterophorus are strongly

limited by water availability (Caño, Escarré & Sans 2007). Each treatment (NW and W) contained one half-

sib per mother plant, with a total of 564 individuals (47 populations x six mother plants/population x two

treatments) randomly distributed across plots within a treatment. Individuals within plots were separated by

75 cm.

Plants were watered during seven weeks after transplanting to minimize mortality and dead plants

were replaced as necessary. The irrigation experiment started in April and ran until October. Plants from the

water treatment received approximately 2212 L/m2 throughout the experiment. Drip irrigation was applied

continuously at a slow rate, and in consequence the soil was never saturated. Precipitation during the course of

the experiment was 535 L/m2; watered plants thus received 413% more water than non-watered plants.

Measurements

Plants were monitored for mortality and phenological stage (vegetative or reproductive) once a week

throughout the experiment. The date of first flowering and the total length of the flowering period were

recorded for each individual, as these characteristics have been related to invasiveness (Pyšek & Richardson

2007). We estimated relative growth rate (RGR) as the increase in plant height, measured at the beginning of

the experiment (week 0) and at weeks 10, 16 and 23. RGR was calculated as the difference in log-transformed

plant height between two consecutive periods divided by the corresponding time interval (first period: 0-10

weeks, second period: 10-16 weeks, and third period: 16-23 weeks). Plant reproductive effort was estimated

by counting the number of flowering and fruiting heads in a plant subsample for each individual in June and

August 2011. Because plant origin and response to irrigation could affect the blooming dynamics, we chose

the highest number of heads counted at either census for each individual as an estimate of head production. To

determine the average number of seeds per head, we counted the number of seeds (achenes) of three heads per

individual plant. Total seed production was estimated by multiplying head production by the average of seeds

per head. Shoot biomass was determined at the end of the experiment (September/October 2011) for all

surviving individuals. Individuals were cut at ground level and leaves were separated from stems. Both

fractions were oven-dried at 65ºC for two to three days and weighed.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Three leaves per plant were collected between September and October to estimate leaf-levels traits.

Fresh leaves were scanned, and foliar area and shape was determined using ImageJ (Schneider, Rasband &

Eliceiri 2012). Leaves were then oven-dried for 72 h at 65 °C and weighed. The SLA was calculated by

dividing leaf area by dry weight. A high SLA is normally associated with higher productivity and

invasiveness (Lake & Leishman 2004) and with shorter life-spans and vulnerability to herbivores and drought

stress (Burke & Grime 1996; Maroco, Pereira & Chaves 2000). Leaf shape was estimated as 4·π· leaf area/leaf

perimetre2 (shape = 1 for a perfect circle). Leaves with more dissected margins are frequently associated with

high evaporation and assimilation rates (Schuepp 1993). Total leaf area was calculated by multiplying leaf dry

weight by SLA.

Leaf N concentration, C/N ratio and C isotopic composition were analysed in 116 individuals (see

Table S1). Leaf N concentration was used as a surrogate for maximum photosynthetic capacity and, hence,

potential growth (Reich, Ellsworth & Walters 1998) whereas δ13C was used as a proxy of water-use efficiency

(Farquhar, Ehleringer & Hubick 1989). All chemical analyses were carried out at the University of California

Davis Stable Isotope Facility (USA) using an IRMS (PDZ Europa ANCA-GSL elemental analyser interfaced

to a PDZ Europa 20-20 isotope ratio mass spectrometer). The relationship between carbon stable isotopes was

expressed in relation to a Pee-Dee Belemnite (PDB) standard. The accuracy of the measurements was

0.015‰.

To characterize the phenotypic plasticity in response to water availability, we calculated a plasticity

index (PI) between half-sibs following Valladares et al. (2000):

PI = [Mean(W) – Mean(NW)] / Max[Mean(W), Mean(NW)]

where Mean(W) and Mean(NW) are trait values of half-sibs growing in the water and non-water treatments,

respectively. PI ranges from 0 (no plasticity) to ±1 (maximum plasticity). A negative PI indicates a higher

mean value under the NW treatment (control) compared with the W treatment.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

STATISTICAL ANALYSES

A generalized binomial mixed model (logit transformation) was used to determine the effects of region,

treatment and their interactions (fixed effects) on plant survival. Mother plants nested within populations and

both crossed with plots were included as random effects. For the quantitative variables measured only once

during the course of the common garden experiment (biomass, SLA, leaf shape, total leaf area, δ13C, N

concentration, C/N, first flowering date, flowering period, seeds per head, number of heads and total number

of seeds) a general linear mixed model was used, including region, water treatment and their interaction as

fixed effects, and the same random structure as before. For RGR, a trait measured repeatedly throughout the

experiment, the model also incorporated time as a fixed factor and individual as an additional random effect

(nested within mother plant). The variables SLA, leaf shape, total leaf area, N concentration, C/N, number of

heads and total number of seeds were normalized by a logarithmic transformation. Statistical analyses of

plasticity were conducted using general linear mixed models with region as fixed effect and population and

plot as random effects.

In a next step of our analysis we asked whether the effects of climate (P/PET) and herbivory

(percentage of predated heads) could explain differences among regions in the studied plant traits. We did that

by fitting additional linear mixed models including region, P/PET and predation as fixed effects. These

models included the interaction between region and climate and between region and predation. As before,

random effects (on the intercept of the model) included mother plant nested within population and both

crossed with plot. Two separate models were fitted: one for control (NW) plants and the other for watered (W)

plants. ANOVA Type I tables corresponding to these models are provided in the Supporting Information (see

Tables S7, S8). In these sequential analyses P/PET was introduced before Region to test whether it explained

the differences across regions obtained in the models presented in the previous paragraph. Note that P/PET

and Predation were not correlated (r2 = 0.002, P > 0.05). The variable P/PET was centred at the mean value

for all populations before the analysis (P/PETcentred= 0.76). The variables SLA, leaf shape, total leaf area, N

concentration, C/N, number of heads, total number of seeds and predation were log transformed to meet

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

normality. Significance for all statistical analyses was accepted at P < 0.05. All models were fitted using the R

software v. 3.1.2 (R Development Core Team 2008) with packages nlme and lme4.

Results

Native and non-native populations of S. pterophorus differed in plant survival, aboveground biomass, total

leaf area, leaf shape and reproductive effort (Fig. 1, see Table S2). Biomass was consistently lower in the non-

native regions, and total leaf area was also lower in Australia and Europe compared with the native range (Fig.

1, see Table S2). Similarly, Australia and Europe had a lower reproductive output (seeds per head, total

number of heads and total number of seeds) although these differences were not always significant (Fig. 1, see

Table S2). Relative to native populations, survival was lower only for Australian plants. No genetic

differences were observed between the native and non-native regions for SLA, δ13C, N concentration, C/N or

phenology (Fig. 1, see Table S2).

Irrigation resulted in a higher biomass and total leaf area, and lower δ13C, C/N and number of seeds

per head (Fig.1, see Table S2). Plant traits showed a plastic response to watering (i.e., plasticity index

different from zero) except for leaf shape, C/N, first flowering date, flowering period and head and seed

production (see Table S4). The response to watering was similar for all regions and most plant traits, except

for phenology in the expanded region, biomass and N and C/N in Australia, and survival in Europe (Fig.1, see

Table S2). However, the plasticity index was not different across regions for any plant trait (the only

exception was flowering period in South Africa expanded range), indicating a similar phenotypic plasticity in

the native and non-native populations (see Table S4). The effects of region and water treatment on the plant

relative growth rate (RGR) were consistent with the pattern observed for biomass (see Fig. S2, Table S3).

The role of climate and herbivory on the differences across regions was evaluated simultaneously in a

statistical model that incorporated summer P/PET and the intensity of predation measured at the population

original areas. In the native region, P/PET was positively related with biomass, total leaf area and first

flowering date and negatively related with leaf shape, N, flowering period and reproductive output variables

(Figs 2 and 3; see Tables S5, S6). The general loss of a significant effect of Region when P/PET was included

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

into the model (see Tables S7, S8) and the fact that the intercepts are similar across regions after accounting

for P/PETcentred (see Tables S5, S6) strongly suggests that the genetic differences in plant size and leaf area

between the native and the non-native regions can be explained by differences in P/PET. Biomass and leaf

area of the non-native populations tended to converge with the native populations with similar climatic

conditions (Figs 2 and 3). In Australia, however, the effect of P/PET on some traits was different to the one

observed for the native region, and was mostly driven by two populations with a much higher P/PET (Figs 2

and 3). Predation was generally unrelated to plant traits, particularly for control (NW) plants (see Tables S7,

S8) and the corresponding model coefficients did not differ among regions (see Tables S5, S6). Finally, leaf

shape and reproductive traits (flowering date, flowering period, number of seeds per head, number of heads

and total number of seeds) were related to P/PET at the native range (Figs 2 and 3, see Tables S5, S6).

However, climate did not completely explain the differences across regions; the overall effect of region for

some of these traits was still significant after removing the variation due to P/PET in the ANOVA analyses,

especially for water treatment (Tables S7, S8).

Discussion

DIFFERENCES BETWEEN NATIVE AND NON-NATIVE POPULATIONS

Our results are consistent with the presence of strong genetically based differences in trait values between the

native and the non-native populations of S. pterophorus. Plants from the non-native areas were smaller and

had lower leaf areas and lower reproductive capacities than plants from the native area. Because the

introduction of S. pterophorus to novel areas is relatively recent (Western Cape ~ 100 years; Australia >70-

100 years; Europe >30 years) (Castells et al. 2013), these results strongly suggest that plant traits can diverge

rapidly after invasion. Moreover, the similar pattern found between Europe and the other two non-native areas

suggests that changes may have occurred early on after the introduction. In contrast, S. pterophorus responded

similarly to watering regardless of their geographic origin. The increased plasticity hypothesis predicts that

plants from invasive populations should be more plastic than plants from the native populations (Richards et

al. 2006). Contrary to this hypothesis, we found no differences in trait plasticity between native and non-

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

native S. pterophorus populations. These results are consistent with previous studies finding that trait values

were more important for determining plant invasibility than trait plasticity (Godoy et al. 2011; Matzek 2012).

CLIMATE DRIVES GEOGRAPHICAL DIVERGENCE IN PLANT TRAITS

S. pterophorus in the native region, in Eastern South Africa, is distributed along an ecological cline of

drought, from the southernmost populations subject to a higher drought stress (lower summer P/PET) to the

northernmost populations growing under wet and cool environments (higher summer P/PET). When plants

from these populations grew under the same environmental conditions in the common garden, we observed a

strong correlation between P/PET at the original sampling locations and most of the measured plant traits.

Plants from drier areas were smaller and had lower leaf area, more dissected leaf margins, earlier blooming,

longer reproductive period and higher seed production compared with plants from more humid areas. These

genetically based trends along a climatic gradient suggest that plants are locally adapted to the conditions in

the native area (Kawecki & Ebert 2004). Indeed, short stature, small size and low leaf area are believed to be

advantageous under dry environmental conditions (Martínez-Vilalta et al. 2009; Hartmann 2011).

Because summer drought is more severe in the three non-native areas (Western South Africa,

Australia and Europe) than in the native range, we hypothesize that climate may have also driven divergence

of vegetative traits after invasion. Several pieces of evidence support this idea. First, the direct effect of region

was no longer significant when P/PETcentred was included into the statistical model (see Tables S7, S8).

Second, the estimated trait values at the mean P/PET were similar across regions (see Tables S5, S6)

indicating that geographical differences in plant traits could be explained by differences in climate. Third, the

direction of the changes across regions was, on average, similar for all of the introduced areas: introduced

plants from Western South Africa, Australia and Europe had lower biomass and leaf area than the native

populations. This pattern is only consistent with a response to similar climatic conditions, as these regions

differed in their introduction time, distance from the source populations, and biotic environment. Finally,

differences in individual-level and leaf-level plant traits after introduction were consistent with the climatic

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

effects within the native region. Moreover, the value of individual-level and leaf-level traits in the non-native

populations tended toward convergence with the native populations under similar climatic conditions, except

for two Australian populations from New South Wales (A01 and A02; see Table S1) that experienced a much

higher P/PET than the rest of populations from the same region.

The role of climate as a main driver for changes in reproductive traits between native and non-native

populations was not as consistent as for vegetative traits. Because plants growing under drier conditions in the

native region had a longer flowering period and a higher head and seed production, we would expect

introduced plants to behave similarly in accordance with their climate. However, non-native plants tended to

have a shorter reproductive season and lower seed production than native plants under similar P/PET

conditions (Fig. 3). These results suggest that in addition to climate, other factors not included in this study

were probably determining geographical differences in reproductive traits.

Genetically based differences between regions could also be caused by non-adaptative events such as

demographic bottlenecks and genetic drift or the plant introduction routes (Keller & Taylor, 2008; Lachmuth

et al., 2011). We cannot reject that neutral events contributed to the geographical divergence, but the

relationship between climate and plant traits within the native region and the convergence between introduced

and native plants from similar climates are indicative of adaptive evolution. Moreover, the similar pattern

observed in all non-native regions suggests that a directional change has occurred in three presumed

independent events. Finally, the genetic similarity across the native and non-native areas obtained by neutral

markers (AFLPs) (Vilatersana et al., unpublished data) shows that the S. pterophorus in Western South

Africa, Australia and Europe comes from multiple introductions spanning a range of climates in the native

area. Thus, the convergence observed in the common garden between the introduced populations and the

native populations with a matching climate cannot be explained solely by the invasion routes.

We acknowledge that our conclusions are limited by the fact that only one common garden located

within the European invaded range was used. Reciprocal common garden experiments have been useful to

reveal the interactions between genetically based plant trait expression and the environmental conditions from

the species’ distributional areas (Williams, Auge & Maron 2008; Colautti, Maron & Barrett 2009). However,

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

we found no biogeographical divergence in trait plasticity in response to water availability (P/PET) and thus,

a priori, we would not expect significant interactions between the relative change in plant traits across regions

and the local environmental conditions in the native and introduced areas.

EVIDENCE AGAINST THE ROLE OF HERBIVORY

The release from natural enemies after invasion has been proposed as a driver for post-invasive evolution

(Crawley, 1987; Blossey & Notzold 1995; Keane & Crawley 2002). One of the most invoked hypotheses

explaining the success of invasive species, the Evolution of Increased Competitive Ability (EICA) hypothesis

(Blossey & Notzold 1995) states that under a lower consumption pressure by specialist herbivores, genotypes

allocating more resources to growth and reproduction and less to chemical defences would be favoured in the

introduced range. In our system, we did not find any relationship between herbivore release and genetically

based plant traits across regions. Contrary to the predictions of the EICA hypothesis, plants from the

introduced populations had lower growth and reproductive output compared with the native populations.

Moreover, this pattern was similar for all non-native regions, even though the release in herbivory was more

intense in Europe than in Australia (Castells et al. 2013).

Experimental support of the EICA remains controversial (Willis, Memmott & Forrester 2000; van

Kleunen & Schmid 2003; Vilà, Gómez & Maron 2003; Jakobs, Weber & Edwards 2004). This lack of

consistent results may occur, at least in part, because comparisons between native and introduced populations

tend to use limited sample sizes and cover only part of the species’ distributional areas. Under these

conditions comparisons between native and introduced populations may not use the appropriate controls,

particularly if the invasion routes are unknown (Bossdorf et al. 2005). Additionally, the release from

herbivores in the invasive range, the first premise of the EICA hypothesis, is rarely evaluated quantitatively.

We overcame most of these limitations by performing a common garden experiment with a large number of

individuals and populations from nearly all of the species’ known distributional range and incorporating data

on herbivore consumption measured in situ on the same mother plants used in the experiment. However, our

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

study was limited by the fact that herbivory was estimated only once on the reproductive parts, so we cannot

discard that herbivory on shoots and roots along the entire plant life cycle could be related to plant divergence

across regions.

In addition, changes in chemical defences after invasion are an important aspect of the EICA hypothesis that

has not been covered in our study. We cannot reject that chemical defences of S. pterophorus, such as

pyrrolizidine alkaloids (Castells et al. 2014), are evolving in response to herbivory independently from

morphological traits.

To our knowledge, this is one of the first studies testing simultaneously, on the same plants, the role of climate

and herbivory as the main drivers of post-invasive evolution. A recent meta-analysis on the North American

invasive plant Lythrum salicaria (Lytharaceae) found that local plant adaptation was driven by both climatic

and biotic effects (Colautti & Barrett 2013). The results obtained here for S. pterophorus are solely consistent

with the role of climate as a driver for plant adaptation to novel environments. Our study adds to the recent

reports of rapid evolution after invasion (Maron et al. 2004; Prentis et al. 2008; Buswell, Moles & Hartley

2011; Colautti & Barrett 2013) by showing that contemporary differentiation may occur in several

independent events. The adaptation of S. pterophorus along a climatic gradient in the native range, together

with multiple introductions in each non-native region, suggests that genotypes pre-adapted to drought (with a

lower growth and leaf area) were favourably selected in the introduced areas, resulting in a rapid geographical

divergence. Although reproductive traits also varied across a climatic cline in the native range, other factors in

addition to drought contributed to their genetic divergence among regions. It remains unresolved whether

genetic changes across regions increased plant fitness as a result of local adaptation, the so called “home site

advantage” (Colautti & Barrett 2013), and whether the potential benefits at the individual level translate to

higher invasion success at the population level. Understanding the mechanisms for rapid differentiation in

response to novel climatic conditions improves our ability not only to explain the dynamics of biological

invasions but also to predict the response of native populations under climate change (Hoffmann & Sgrò

2011).

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Acknowledgements

We thank Miriam Cabezas, Maria Morante, Pere Losada, Guillem Esparza, Xavier Sans and Jose Manuel

Blanco-Moreno for technical assistance. This research was conducted thanks to the financial support provided

to E.C. by Ministerio de Ciencia e Innovación (Spain) (GCL2008-02421/BOS) and Ministerio de Economía y

Competitividad (Spain) (GCL2011-29205). E.C. and J.M.V. belong to the research group “Response of

ecosystems to climate change and environmental gradients” funded by Generalitat de Catalunya (Catalonia)

(2014 SGR-453). We thank an anonymous reviewer for her/his exhaustive contributions which improved the

quality of the final manuscript. The authors declare that they have no conflicts of interests.

Data Accessibility

Data deposited in the repository of Universitat Autònoma de Barcelona: https://ddd.uab.cat/record/131539

References

Andreu, J., Pino, J., Basnou, C., Guardiola, M. & Ordóñez, J.L. (2012) Les Espècies Exòtiques de

Catalunya. Resum Del Projecte EXOCAT 2012 (ed Departament d’Agricultura Ramaderia

Pesca Alimentació i Medi Natural). Generalitat de Catalunya.

Blossey, B. & Notzold, R. (1995) Evolution of increased competitive ability in invasive

nonindigenous plants: a hypothesis. The Journal of Ecology, 83, 887.

Bossdorf, O., Auge, H., Lafuma, L., Rogers, W.E., Siemann, E. & Prati, D. (2005) Phenotypic and

genetic differentiation between native and introduced plant populations. Oecologia, 144, 1–11.

Burke, M.J.W. & Grime, J.P. (1996) An experimental study of plant community invasibility.

Ecology, 77, 776–790.

Buswell, J.M., Moles, A.T. & Hartley, S. (2011) Is rapid evolution common in introduced plant

species? Journal of Ecology, 99, 214–224.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Caño, L., Escarré, J. & Sans, F.X. (2007) Factors affecting the invasion success of Senecio

inaequidens and S.pterophorus in Mediterranean plant communities. Journal of Vegetation

Science, 18, 279–286.

Castells, E., Morante, M., Blanco-Moreno, J.M., Sans, F.X., Vilatersana, R. & Blasco-Moreno, A.

(2013) Reduced seed predation after invasion supports enemy release in a broad

biogeographical survey. Oecologia 173, 1397-1409.

Castells, E., Mulder,P.P.J., Pérez-Trujillo, M. (2014) Diversity of pyrrolizidine alkalodis in native

and invasive Senecio pterophorus (Asteraceae): implications for toxicity. Phytochemistry 108,

137-146.

Colautti, R.I. & Barrett, S.C.H. (2013) Rapid adaptation to climate facilitates range expansion of an

invasive plant. Science, 342, 364–366.

Colautti, R.I., Maron, J.L. & Barrett, S.C.H. (2009) Common garden comparisons of native and

introduced plant populations: latitudinal clines can obscure evolutionary inferences.

Evolutionary Applications, 2, 187–199.

Crawley, M. J. (1987) What makes a community invasible? Colonization, succession and stability

(eds A. J. Gray, M. J. Crawley & P. J. Edwards), pp. 429-853. Blackwell Scientific

Publications, Oxford.

Farquhar, G.D., Ehleringer, J.R. & Hubick, K.T. (1989) Carbon isotope discrimination and

photosynthesis. Annual Reviews os Plant Physiology and Plant Molecular Biology, 40, 503–

537.

Godoy, O., Valladares, F. & Castro-Díez, P. (2011) Multispecies comparison reveals that invasive

and native plants differ in their traits but not in their plasticity. Functional Ecology, 25, 1248–

1259.

Hartmann, H. (2011) Will a 385 million year-struggle for light become a struggle for water and for

carbon? – How trees may cope with more frequent climate change-type drought events. Global

Change Biology, 17, 642–655.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis, A. (2005) Very high resolution

interpolated climate surfaces for global land areas. International Journal of Climatology, 25,

1965–1978.

Hilliard, O.M. (1977) Compositae in Natal. University of Natal Press, Pietermaritzburg.

Hoffmann, A.A. & Sgrò, C.M. (2011) Climate change and evolutionary adaptation. Nature, 470,

479–485.

Jakobs, G., Weber, E. & Edwards, P.J. (2004) Introduced plants of the invasive Solidago gigantea

(Asteraceae) are larger and grow denser than conspecifics in the native range. Diversity and

Distributions, 10, 11–19.

Kawecki, T.J. & Ebert, D. (2004) Conceptual issues in local adaptation. Ecology letters, 7, 1225–

1241.

Keane, R.M. & Crawley, M.J. (2002) Exotic plant invasions and the enemy release hypothesis.

Trends in Ecology & Evolution, 17, 164–170.

Keller, S. R. & Taylor, D. R. (2008) History, chance and adaptation during biological invasion:

separating stochastic phenotypic evolution from response to selection. Ecology Letters, 11,

852–866.

Kolar, C.S. & Lodge, D.M. (2001) Progress in invasion biology: predicting invaders. Trends in

Ecology & Evolution, 16, 199–204.

Lachmuth, S., Durka, W. & Schurr, F.M. (2011) Differentiation of reproductive and competitive

ability in the invaded range of Senecio inaequidens: the role of genetic Allee effects, adaptive

and nonadaptive evolution. New Phytologist, 192, 529–541.

Lake, J.C. & Leishman, M.R. (2004) Invasion success of exotic plants in natural ecosystems: the role

of disturbance, plant attributes and freedom from herbivores. Biological Conservation, 117,

215–226.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Mack, R.N., Simberloff, D., Lonsdale, W.M., Evans, H., Clout, M. & Bazzaz, F.A. (2000) Biotic

invasions: causes, epidemiology, global consequences, and control. Ecological Applications, 10,

689–710.

Maroco, J.P., Pereira, J.S. & Chaves, M.M. (2000) Growth, photosynthesis and water-use efficiency

of two C4 Sahelian grasses subjected to water deficits. Journal of Arid Environments, 45, 119–

137.

Maron, J.L., Vilà, M., Bommarco, R., Elmendorf, S. & Beardsley, P. (2004) Rapid evolution of an

invasive plant. Ecological Monographs, 74, 261–280.

Martínez-Vilalta, J., Lopez, B.C., Adell, N., Badiella, L. & Ninyerola, M. (2008) Twentieth century

increase of Scots pine radial growth in NE Spain shows strong climate interactions. Global

Change Biology, 14, 2868-2881.

Martínez-Vilalta, J., Cochard, H., Mencuccini, M., Sterck, F., Herrero, A., Korhonen, J.F.J., Llorens,

P., Nikinmaa, E., Nolè, A., Poyatos, R., Ripullone, F., Sass-Klaassen, U. & Zweifel, R. (2009)

Hydraulic adjustment of Scots pine across Europe. New Phytologist, 184, 353–364.

Matzek, V. (2012) Trait Values, Not Trait Plasticity, Best Explain Invasive Species’ Performance in

a Changing Environment. PloS one, 7, e48821.

Meteorological Service of Catalonia (2015), URL www.meteo.cat [accessed 21 January 2015]

Moles, A.T., Warton, D.I., Warman, L., Swenson, N.G., Laffan, S.W., Zanne, A.E., Pitman, A.,

Hemmings, F.A. & Leishman, M.R. (2009) Global patterns in plant height. Journal of Ecology,

97, 923–932.

Ninyerola, M., Pons, X., Roure, J.M., Martin Vide, J., Raso, J.M. & Clavero, P. (2003) Atles

Climàtic Digital de Catalunya. Servei Meteorològic de Catalunya and Departament de Medi

Ambient, Generalitat de Catalunya.

Parsons, W.T. & Cuthbertson, E.G. (1992) Noxious Weeds of Australia. Csiro Pub.

Prentis, P.J., Wilson, J.R.U., Dormontt, E.E., Richardson, D.M. & Lowe, A.J. (2008) Adaptive

evolution in invasive species. Trends in plant science, 13, 288–94.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Pyšek, P. & Richardson, D.M. (2007) Traits associated with invasiveness in alien pPlants: Where do

we stand? Biological Invasions (ed W. Nentwig), pp 97-125. Springer, Berlin.

R Development Core Team (2008). R: A language and environment for statistical computing. R

Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org

[accessed 21 January 2015]

Reich, P.B., Ellsworth, D.S. & Walters, M.B. (1998) Leaf structure (specific leaf area) modulates

photosynthesis-nitrogen relations: evidence from within and across species and functional

groups. Functional Ecology, 12, 948–958.

Richards, C.L., Bossdorf, O., Muth, N.Z., Gurevitch, J. & Pigliucci, M. (2006) Jack of all trades,

master of some? On the role of phenotypic plasticity in plant invasions. Ecology letters, 9, 981–

93.

Sakai, A.K., Allendorf, F.W., Holt, J.S., Lodge, D.M., Molofsky, J., With, K.A., Baughman, S.,

Cabin, R.J., Cohen, J.E., Ellstrand, N.C., McCauley, D.E., O’Neil, P., Parker, I.M., Thompson,

J.N. & Weller, S.G. (2001) The population biology of invasive species. Annual Reviews of

Ecology and Systematics, 32, 305–332.

Schneider, C.A., Rasband, W.S., Eliceiri, K.W. (2012) NIH Image to ImageJ: 25 years of image

analysis. Nature Methods, 9, 671-675.

Schuepp, P.H. (1993) Tansley Review No. 59 Leaf boundary layers. New Phytologist, 125, 477–507.

Soil Survey Staff (2010) Keys to Soil Taxonomy, 11th ed. USDA-Natural Resources Conservation

Service, Washington, DC.

Sol, D., Maspons, J., Vall-llosera, M., Bartomeus, I., García-Peña, G.E., Piñol, J. & Freckleton, R.P.

(2012) Unraveling the life history of succesful invaders. Science, 337, 580–583.

Thornthwaite, C. W. (1948) An approach toward a rational classification of climate. Geographical

Review 38, 55–94.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Valladares, F., Wright, S.J., Lasso, E., Kitajima, K. & Pearcy, R.W. (2000) Plastic phenotypic

response to light of 16 congeneric shrubs from a Panamanian rainforest. Ecology, 81, 1925–

1936.

Van Kleunen, M. & Schmid, B. (2003) No evidence for an evolutionary increased competitive

ability in an invasive plant. Ecology, 84, 2816–2823.

Vanderhoeven, S., Brown, C.S., Tepolt, C.K., Tsutsui, N.D., Vanparys, V., Atkinson, S., Mahy, G. &

Monty, A. (2010) Perspective: Linking concepts in the ecology and evolution of invasive plants:

network analysis shows what has been most studied and identifies knowledge gaps.

Evolutionary Applications, 3, 193–202.

Vilà, M., Gómez, A. & Maron, J.L. (2003) Are alien plants more competitive than their native

conspecifics? A test using Hypericum perforatum L. Oecologia, 137, 211–215.

Westoby, M., Falster, D.S., Moles, A.T., Vesk, P.A. & Wright, I.J. (2002) Plant ecological strategies:

some leading dimensions of variation between species. Annual Review of Ecology and

Systematics, 33, 125–159.

Williams, J.L., Auge, H. & Maron, J.L. (2008) Different gardens, different results: native and

introduced populations exhibit contrasting phenotypes across common gardens. Oecologia, 157,

239–48.

Williamson, M. (1996) Biological Invasions. Chapman & Hall, London.

Willis, A.J. & Blossey, B. (1999) Benign environments do not explain the increased vigour of non-

indigenous plants: a cross-continental transplant experiment. Biocontrol Science and

Technology, 9, 567–577.

Willis, A.J., Memmott, J. & Forrester, R.I. (2000) Is there evidence for the post-invasion evolution

of increased size among invasive plant species? Ecology letters, 3, 275–283.

Supporting Information

Additional Supporting information may be found in the online version of this article:

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Table S1. Characteristics of the populations of Senecio pterophorus used in the common garden

experiment

Table S2. Estimates and significance of the effects of region and treatment on individual-level traits,

leaf-level traits and reproductive traits

Table S3. Estimates and significance of the effects of region and treatment on the relative growth

rate

Table S4. Estimates and significance of the plasticity index in response to a water treatment

Table S5. Estimates and significance of the effects of region, drought index and herbivory on

individual-level traits, leaf-level traits and reproductive traits of plants growing under a control (non-

watered) treatment

Table S6. Estimates and significance of the effects of region, drought index and herbivory on

individual-level traits, leaf-level traits and reproductive traits of plants growing under a water

treatment

Table S7. ANOVA Type I table corresponding to the linear mixed model presented in Table S5

Table S8. ANOVA Type I table corresponding to the linear mixed model presented in Table S6

Fig. S1. Monthly temperature and precipitation at the S. pterophorus sampling locations averaged by

region

Fig. S2. Plant height of S. pterophorus from the native region compared to three introduced regions

in a common garden under a water treatment

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Table 1. Characteristics of the Senecio pterophorus populations used in the common garden experiment,

averaged by region in the native, expanded and introduced ranges (Mean ± SE). Different letters indicate

significant differences between regions by a Tukey post-hoc contrast in a t-test

Region Plant Status

Populations and individuals

Elevation*

(m)

Mean Annual Temperature*

(ºC)

Mean Annual Precipitation*

(mm)

Summer P/PET†

Predation*

(% damaged heads)

South Africa

Native 18 pop, 107 ind 792.7 ± 96.3a 16.6 ± 0.2a 746.2 ± 31.8 1.27 ± 0.11a 25.2 ± 1.6a

Expanded 5 pop, 29 ind 133.0 ± 56.1b 16.1 ± 0.4ab 856.4 ± 86.5 0.36 ± 0.04b 33.4 ± 2.8a

Australia Introduced

12 pop, 70 ind 140.7 ± 46.9b 15.1 ± 0.4b 754.4 ± 63.0 0.51 ± 0.09b 15.4 ± 1.5b

Europe Introduced

12 pop, 72 ind 244.5 ± 46.9b 15.3 ± 0.2b 667.0 ± 19.9 0.37 ± 0.02b 0.2 ± 0.1c

* Data obtained from Castells et al. 2013

†Ratio of precipitation to potential evapotranspiration during summer (December-February in the southern hemisphere and June-August in the northern hemisphere).

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Firs

t flo

wer

ing

(day

s)

80

100

120

140

160

Flo

wer

ing

perio

d (d

ays)

0

20

40

60

80

100

Num

. hea

ds (

x103 )

0

1

2

3

4

See

ds /

head

60

80

100

120

Num

. See

ds (

x105 )

0

1

2

3

4

5

C/N

10

15

20

25

30

N (

%)

1.5

2.0

2.5

3.0

3.5

4.0

δ13C

-32

-30

-28

-26

Leaf

sha

pe

0.05

0.10

0.15

0.20

0.25

0.30

0.35

Leaf

are

a (d

m2 )

0

200

400

600

800S

LA (

cm2

/ g D

M)

75

100

125

150

175

200

225

Ste

m b

iom

ass

(g D

M)

0

250

500

750

1000

1250

1500

Leaf

bio

mas

s (g

DM

)

0

100

200

300

400

500

Sho

ot b

iom

ass

(g D

M)

0

500

1000

1500

2000

Sur

viva

l (%

)

40

60

80

100Non-wateredWatered

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(i)

(k)

(l)

(m)

(n)

(j) (o)

SANative

SAExp.

AUS EUR SANative

SAExp.

AUS EUR SANative

SAExp.

AUS EUR

(h)

*

**

***

**

**

**

*

**

**

**

**

*** **

**

**

*** *

****

*

*

*

**

**

**

**

*

*

*

Fig. 1 Individual-level traits (a-d), leaf-level traits (e-j) and reproductive traits (k-o) of S. pterophorus from

the South Africa native range (SA Native), South Africa expanded range (SA Exp.) and two introduced

regions in Australia (AUS) and Europe (EUR), growing in a common garden experiment without irrigation

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

(non-watered; light grey) and with irrigation (watered; dark grey). Means ± SE are shown for percentage

survival. The boxplots for the other variables show the 25th and 75th percentiles (box limits), the median (inner

line), and the 10th and 90th (below and above whiskers, respectively). Statistical significance corresponding to

the linear mixed models of Table S2 is shown as *P < 0.05, **P< 0.01 and ***P < 0.001. The reference level

(intercept) is South Africa native range and non-watered treatment. Asterisks on top of a non-watered

treatment in the non-native regions (South Africa expanded, Australia and Europe) indicate significant

differences in this control treatment between each region and South Africa native range. Significant

differences between regions for the water treatment are not shown for simplicity. Asterisks on top of a

horizontal bar in South Africa native range indicate a significant effect of water treatment in this region.

Finally, asterisks on top of a horizontal bar in the non-native regions indicate differences in the treatment

effect (Region:Treatment) between each of these regions and South Africa native range.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

P/PET (absolute)

0.0 0.5 1.0 1.5 2.0

P/PET (absolute)0.0 0.5 1.0 1.5 2.0

NW treatment

Sho

ot b

iom

ass

(g D

M)

0

250

500

750

1000

1250

1500

Leaf

bio

mas

s (g

DM

)

0

100

200

300

400

W treatment

South Africa-nativeSouth Africa-expandedAustraliaEurope

(a)

(b) (g)

(f)

Ste

m b

iom

ass

(g D

M)

0

200

400

600

800

1000

1200

Tot

al le

af a

rea

(cm

2 )

e7

e8

e9

e10

e11

e12

P/PET (centred)-0.5 0.0 0.5 1.0

Leaf

sha

pe

e-3

e-2

e-1

P/PET (centred)-0.5 0.0 0.5 1.0

(c)

(d)

(e)

(h)

(i)

(j)

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Fig. 2 Relation between drought index and individual-level and leaf-level traits of S. pterophorus from the

native region in South Africa, the expanded region in South Africa and the introduced regions in Australia and

Europe. Drought index was expressed as the ratio of summer precipitation to potential evapotranspiration -

P/PET- centred at the average for all populations. Plants were grown in a common garden experiment without

irrigation (NW treatment) (a-e) and with irrigation (W treatment) (f-j). Each dot represents a population

average. Depicted lines represent the relationships as obtained from the coefficients of the corresponding

linear mixed models (see Tables S5, S6). For clarity, relationships are only shown for the native region (solid

line) and for any region when the slope of the relationship was significantly different from South Africa-

native (in our case only Australia, dashed line). An additional x-axis with the absolute value of P/PET is

shown as a reference.

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

P/PET (absolute)

0.0 0.5 1.0 1.5 2.0

P/PET (absolute)

0.0 0.5 1.0 1.5 2.0

Num

. hea

ds (

Ln)

5

6

7

8

9

NW treatment

Firs

t flo

wer

ing

date

(da

ys)

80

100

120

140

160

180

Flo

wer

ing

perio

d (d

ays)

10

30

50

70

90

W treatment

South Africa-nativeSouth Africa-expandedAustraliaEurope

(a)

(b) (g)

(f)

See

ds /

head

40

60

80

100

120

P/PET (centred)-0.5 0.0 0.5 1.0

Tot

al n

um. s

eeds

(Ln

)

10

11

12

13

14

P/PET (centred)-0.5 0.0 0.5 1.0

(c)

(d)

(e)

(h)

(i)

(j)

Acc

epte

d A

rtic

le

This article is protected by copyright. All rights reserved.

Fig. 3 Relation between drought index and reproductive traits of S. pterophorus from the native region in

South Africa, the expanded region in South Africa and the introduced regions in Australia and Europe.

Drought index was expressed as the ratio of summer precipitation to potential evapotranspiration -P/PET-

centred at the average for all populations. Plants were grown in a common garden experiment without

irrigation (NW treatment) (a-e) and with irrigation (W treatment) (f-j). Each dot represents a population

average. Depicted lines represent the relationships as obtained from the coefficients of the corresponding

linear mixed models (see Tables S5, S6). For clarity, relationships are only shown for the native region (solid

line) and for any region when the slope of the relationship was significantly different from South Africa-

native (in our case only Australia for some traits, dashed line). An additional x-axis with the absolute value of

P/PET is shown as a reference.